Time Series Credit Information Makes Highly Desired Metrics Feasible
The production of relevant credit industry performance metrics from U.S. consumer reports has been a capability generally neglected by the top U.S. CRAs. For decades, especially since “the Great Recession”, benefactors of credit industry performance measures have attempted to secure industry level loss curves and other meaningful portfolio and vintage performance metrics from consumer credit reports with little success. Primarily attributed to data limitations, the generation of highly focused performance metrics is cost prohibitive, complex and arduous process, until now.
Elimination of Credit Archive Processing Makes Loss Metrics Economical
Until recently, the creation of the aforementioned metrics required processing vast amounts of consumer credit files from multiple archived credit bureau snapshots which then needed to be linked together to create a time series at the consumer level. To produce time series information for balance and credit limit amounts accounts of interest where needed to be identified from each period within the time series consumer credit file and piecemealed together using complex account matching programs. This was a lengthy and expensive process. For example, the creation of a 48 month loss curve on a small sample of consumers would require matching and interrogating 48 monthly archives to produce basic loss curve components, easily costing the end user hundreds of thousands of dollars. The recent availability of account level balance, credit limit and payment time series amounts on U.S. consumer credit reports eliminates the need to access archived credit files and the necessity to link account level histories across a time horizon.
Acute Credit Analysts Can Fill in the Missing Pieces
Although the process to obtain time series consumer credit information on a sample of consumers or account holders has been simplified at each of the U.S. CRAs there are a few missing ingredients to create a standardized, turnkey process to generate credit industry loss metrics. Analysts with an intimate understanding of consumer credit report information are needed to develop rules and program software that identifies, selects and organizes (or as I prefer to say “calendarizes”) tradeline level information by account origination date, delinquency status and product or industry classification which is then summarized to yield meaningful performance metrics over a portfolio’s or vintage’s life cycle.
Who will take the Initiative?
The intimate understanding of consumer credit reports necessary to develop the rules and software to accurately identify, select and organize tradeline level information, which is the key to unlocking this the new rich data source, is understood by a handful of stakeholders. Analysts at CRAs, lenders of significant size and third party analytic companies have the knowledge and skills to develop the rules and software necessary to compute these statistics, each having a different motivation to perform this function. Lenders stand to benefit from having their own proprietary performance software; customized software will better meet their own unique needs and would minimize data fees to CRAs and prevent third party analytic royalty fees. Third party analytic companies benefit by enhancing their product and service capabilities. But CRAs may have the most motivation.
A Defining Moment for U.S CRAs
It appears that CRAs stand to benefit the most from creating a “soup to nuts” standardized process to leverage this new information. The offering of industry and lender specific performance loss metrics is a golden opportunity for U.S. CRAs to further entrench themselves as an analytic and value added service provider to a broader audience of stakeholders that find value in summarized portfolio and industry performance measure. An offering of this nature protects and enhances future data revenue streams and stands to increase future analytic revenue streams.
Because lenders and third party analytic providers, that support lender activities, have access to anonymous consumer credit information either party could easily develop proprietary software to create this same information. Unless U.S.CRAs act quickly, their idleness may relegate themselves lower in the “value added food chain” as merely data providers, stifling the possibility of future revenue streams.
About the Author: Chet Wiermanski is one of BIIA’s contributing editors writing on the subjects of credit scoring and decision systems. He is a Visiting Scholar at the Federal Reserve Bank of Philadelphia researching new applications of consumer credit report information. Additionally, Chet is Managing Director of Aether Analytics which specializes on leveraging hidden data sequences and time series components within consumer credit information typically ignored by traditional credit bureau based solutions. Previously Chet was the Global Chief Scientist at TransUnion LLC. Holding a variety of positions within TransUnion, during his tenure, between July 1997 and February 2012, he was responsible for identifying, evaluating and developing new technology platforms involving alternative data sources, predictive modeling, econometric forecasting and related consulting services.